AutoML Leaderboard
| Best model | name | model_type | metric_type | metric_value | train_time | single_prediction_time |
|---|---|---|---|---|---|---|
| 1_Linear | Linear | accuracy | 0.930131 | 30.25 | 0.0319 | |
| 2_Default_LightGBM | LightGBM | accuracy | 0.962882 | 27.45 | 0.017 | |
| 3_Default_Xgboost | Xgboost | accuracy | 0.962882 | 27.61 | 0.0229 | |
| 4_Default_CatBoost | CatBoost | accuracy | 0.965066 | 27.32 | 0.0153 | |
| 5_Default_NeuralNetwork | Neural Network | accuracy | 0.934498 | 25.89 | 0.0289 | |
| 6_Default_RandomForest | Random Forest | accuracy | 0.921397 | 30.07 | 0.1462 | |
| 11_LightGBM | LightGBM | accuracy | 0.938865 | 27.36 | 0.0119 | |
| 7_Xgboost | Xgboost | accuracy | 0.949782 | 28.25 | 0.0123 | |
| 15_CatBoost | CatBoost | accuracy | 0.965066 | 28.26 | 0.019 | |
| 19_RandomForest | Random Forest | accuracy | 0.960699 | 30.18 | 0.1544 | |
| 23_NeuralNetwork | Neural Network | accuracy | 0.943231 | 27.74 | 0.0377 | |
| 12_LightGBM | LightGBM | accuracy | 0.967249 | 28.78 | 0.0111 | |
| 8_Xgboost | Xgboost | accuracy | 0.960699 | 28.64 | 0.0159 | |
| 16_CatBoost | CatBoost | accuracy | 0.962882 | 29.26 | 0.0209 | |
| 20_RandomForest | Random Forest | accuracy | 0.923581 | 31.19 | 0.1406 | |
| 24_NeuralNetwork | Neural Network | accuracy | 0.941048 | 28.28 | 0.0303 | |
| 13_LightGBM | LightGBM | accuracy | 0.967249 | 28.2 | 0.0123 | |
| 9_Xgboost | Xgboost | accuracy | 0.89738 | 28.24 | 0.0169 | |
| 17_CatBoost | CatBoost | accuracy | 0.967249 | 29.14 | 0.0165 | |
| 21_RandomForest | Random Forest | accuracy | 0.899563 | 30.84 | 0.1379 | |
| 25_NeuralNetwork | Neural Network | accuracy | 0.936681 | 28.97 | 0.0396 | |
| 14_LightGBM | LightGBM | accuracy | 0.967249 | 30.63 | 0.0116 | |
| 10_Xgboost | Xgboost | accuracy | 0.502183 | 29.14 | 0.0131 | |
| 18_CatBoost | CatBoost | accuracy | 0.965066 | 29.87 | 0.0129 | |
| 22_RandomForest | Random Forest | accuracy | 0.914847 | 31.49 | 0.1518 | |
| 26_NeuralNetwork | Neural Network | accuracy | 0.943231 | 28.93 | 0.0357 | |
| 14_LightGBM_GoldenFeatures | LightGBM | accuracy | 0.962882 | 31.65 | 0.0362 | |
| 17_CatBoost_GoldenFeatures | CatBoost | accuracy | 0.958515 | 30.24 | 0.0339 | |
| 13_LightGBM_GoldenFeatures | LightGBM | accuracy | 0.956332 | 30.79 | 0.0323 | |
| 27_LightGBM | LightGBM | accuracy | 0.965066 | 31.37 | 0.0126 | |
| 28_CatBoost | CatBoost | accuracy | 0.967249 | 30.62 | 0.0184 | |
| 29_LightGBM | LightGBM | accuracy | 0.962882 | 30 | 0.0174 | |
| 30_LightGBM | LightGBM | accuracy | 0.969432 | 30.53 | 0.0111 | |
| 31_CatBoost | CatBoost | accuracy | 0.969432 | 30.15 | 0.0143 | |
| 32_Xgboost | Xgboost | accuracy | 0.949782 | 31.02 | 0.0128 | |
| 33_RandomForest | Random Forest | accuracy | 0.960699 | 32.4 | 0.1572 | |
| 34_RandomForest | Random Forest | accuracy | 0.941048 | 32.84 | 0.1487 | |
| 35_Xgboost | Xgboost | accuracy | 0.954148 | 31.76 | 0.0137 | |
| 36_NeuralNetwork | Neural Network | accuracy | 0.91048 | 31.56 | 0.0285 | |
| 37_NeuralNetwork | Neural Network | accuracy | 0.91048 | 30.74 | 0.0369 | |
| 38_NeuralNetwork | Neural Network | accuracy | 0.943231 | 31.05 | 0.029 | |
| 39_NeuralNetwork | Neural Network | accuracy | 0.938865 | 30.66 | 0.049 | |
| 40_RandomForest | Random Forest | accuracy | 0.958515 | 33.89 | 0.1399 | |
| 41_RandomForest | Random Forest | accuracy | 0.91048 | 33.33 | 0.1298 | |
| 42_CatBoost | CatBoost | accuracy | 0.969432 | 32.63 | 0.0163 | |
| 43_LightGBM | LightGBM | accuracy | 0.967249 | 31.89 | 0.0117 | |
| 44_CatBoost | CatBoost | accuracy | 0.967249 | 31.59 | 0.0151 | |
| 45_CatBoost | CatBoost | accuracy | 0.967249 | 31.68 | 0.0192 | |
| 46_LightGBM | LightGBM | accuracy | 0.965066 | 31.68 | 0.0136 | |
| 47_Xgboost | Xgboost | accuracy | 0.962882 | 32.56 | 0.0131 | |
| the best | Ensemble | Ensemble | accuracy | 0.973799 | 2.86 | 0.0381 |
AutoML Performance
AutoML Performance Boxplot
Features Importance
Spearman Correlation of Models
Summary of 10_Xgboost
Extreme Gradient Boosting (Xgboost)
- n_jobs: -1
- objective: binary:logistic
- eta: 0.15
- max_depth: 8
- min_child_weight: 50
- subsample: 0.6
- colsample_bytree: 0.6
- eval_metric: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
28.3 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.693166 | nan |
| auc | 0.501745 | nan |
| f1 | 0.666667 | 0.447816 |
| accuracy | 0.502183 | 0.497573 |
| precision | 0.501818 | 0.503261 |
| recall | 1 | 0.447816 |
| mcc | 0.005472 | 0.497573 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.693166 | nan |
| auc | 0.501745 | nan |
| f1 | 0.61745 | 0.497573 |
| accuracy | 0.502183 | 0.497573 |
| precision | 0.501362 | 0.497573 |
| recall | 0.803493 | 0.497573 |
| mcc | 0.005472 | 0.497573 |
Confusion matrix (at threshold=0.497573)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 46 | 183 |
| Labeled as 1 | 45 | 184 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 11_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 15
- learning_rate: 0.05
- feature_fraction: 0.8
- bagging_fraction: 0.5
- min_data_in_leaf: 50
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
26.6 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.333978 | nan |
| auc | 0.964922 | nan |
| f1 | 0.940171 | 0.494684 |
| accuracy | 0.938865 | 0.494684 |
| precision | 1 | 0.863332 |
| recall | 1 | 0.0302798 |
| mcc | 0.878567 | 0.494684 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.333978 | nan |
| auc | 0.964922 | nan |
| f1 | 0.940171 | 0.494684 |
| accuracy | 0.938865 | 0.494684 |
| precision | 0.920502 | 0.494684 |
| recall | 0.960699 | 0.494684 |
| mcc | 0.878567 | 0.494684 |
Confusion matrix (at threshold=0.494684)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 210 | 19 |
| Labeled as 1 | 9 | 220 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 12_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 63
- learning_rate: 0.2
- feature_fraction: 0.5
- bagging_fraction: 1.0
- min_data_in_leaf: 10
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
28.0 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.234848 | nan |
| auc | 0.975906 | nan |
| f1 | 0.964758 | 0.506583 |
| accuracy | 0.965066 | 0.506583 |
| precision | 1 | 0.983528 |
| recall | 1 | 6.92688e-05 |
| mcc | 0.930273 | 0.506583 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.234848 | nan |
| auc | 0.975906 | nan |
| f1 | 0.964758 | 0.506583 |
| accuracy | 0.965066 | 0.506583 |
| precision | 0.973333 | 0.506583 |
| recall | 0.956332 | 0.506583 |
| mcc | 0.930273 | 0.506583 |
Confusion matrix (at threshold=0.506583)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 223 | 6 |
| Labeled as 1 | 10 | 219 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 13_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 63
- learning_rate: 0.05
- feature_fraction: 0.9
- bagging_fraction: 1.0
- min_data_in_leaf: 20
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
27.3 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.239979 | nan |
| auc | 0.979482 | nan |
| f1 | 0.964444 | 0.503711 |
| accuracy | 0.965066 | 0.503711 |
| precision | 0.994083 | 0.757501 |
| recall | 1 | 0.00133263 |
| mcc | 0.930699 | 0.503711 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.239979 | nan |
| auc | 0.979482 | nan |
| f1 | 0.964444 | 0.503711 |
| accuracy | 0.965066 | 0.503711 |
| precision | 0.9819 | 0.503711 |
| recall | 0.947598 | 0.503711 |
| mcc | 0.930699 | 0.503711 |
Confusion matrix (at threshold=0.503711)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 225 | 4 |
| Labeled as 1 | 12 | 217 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 13_LightGBM_GoldenFeatures
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 63
- learning_rate: 0.05
- feature_fraction: 0.9
- bagging_fraction: 1.0
- min_data_in_leaf: 20
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.9 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.302545 | nan |
| auc | 0.97378 | nan |
| f1 | 0.955947 | 0.499862 |
| accuracy | 0.956332 | 0.499862 |
| precision | 1 | 0.866791 |
| recall | 1 | 0.00305654 |
| mcc | 0.912803 | 0.499862 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.302545 | nan |
| auc | 0.97378 | nan |
| f1 | 0.955947 | 0.499862 |
| accuracy | 0.956332 | 0.499862 |
| precision | 0.964444 | 0.499862 |
| recall | 0.947598 | 0.499862 |
| mcc | 0.912803 | 0.499862 |
Confusion matrix (at threshold=0.499862)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 221 | 8 |
| Labeled as 1 | 12 | 217 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 14_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 95
- learning_rate: 0.05
- feature_fraction: 1.0
- bagging_fraction: 1.0
- min_data_in_leaf: 10
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.9 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.176663 | nan |
| auc | 0.977251 | nan |
| f1 | 0.964758 | 0.493538 |
| accuracy | 0.965066 | 0.493538 |
| precision | 1 | 0.948169 |
| recall | 1 | 0.00591764 |
| mcc | 0.930273 | 0.493538 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.176663 | nan |
| auc | 0.977251 | nan |
| f1 | 0.964758 | 0.493538 |
| accuracy | 0.965066 | 0.493538 |
| precision | 0.973333 | 0.493538 |
| recall | 0.956332 | 0.493538 |
| mcc | 0.930273 | 0.493538 |
Confusion matrix (at threshold=0.493538)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 223 | 6 |
| Labeled as 1 | 10 | 219 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 14_LightGBM_GoldenFeatures
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 95
- learning_rate: 0.05
- feature_fraction: 1.0
- bagging_fraction: 1.0
- min_data_in_leaf: 10
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.7 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.249633 | nan |
| auc | 0.973218 | nan |
| f1 | 0.96 | 0.482912 |
| accuracy | 0.960699 | 0.482912 |
| precision | 0.995025 | 0.683605 |
| recall | 1 | 0.00615977 |
| mcc | 0.92196 | 0.482912 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.249633 | nan |
| auc | 0.973218 | nan |
| f1 | 0.96 | 0.482912 |
| accuracy | 0.960699 | 0.482912 |
| precision | 0.977376 | 0.482912 |
| recall | 0.943231 | 0.482912 |
| mcc | 0.92196 | 0.482912 |
Confusion matrix (at threshold=0.482912)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 224 | 5 |
| Labeled as 1 | 13 | 216 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 15_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.05
- depth: 8
- rsm: 0.8
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
27.5 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.455814 | nan |
| auc | 0.986261 | nan |
| f1 | 0.962138 | 0.493138 |
| accuracy | 0.962882 | 0.493138 |
| precision | 1 | 0.602267 |
| recall | 1 | 0.163765 |
| mcc | 0.92648 | 0.493138 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.455814 | nan |
| auc | 0.986261 | nan |
| f1 | 0.962138 | 0.493138 |
| accuracy | 0.962882 | 0.493138 |
| precision | 0.981818 | 0.493138 |
| recall | 0.943231 | 0.493138 |
| mcc | 0.92648 | 0.493138 |
Confusion matrix (at threshold=0.493138)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 225 | 4 |
| Labeled as 1 | 13 | 216 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 16_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.1
- depth: 8
- rsm: 1.0
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
28.5 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.300488 | nan |
| auc | 0.977403 | nan |
| f1 | 0.960352 | 0.482133 |
| accuracy | 0.960699 | 0.482133 |
| precision | 1 | 0.847893 |
| recall | 1 | 0.0100179 |
| mcc | 0.92196 | 0.504025 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.300488 | nan |
| auc | 0.977403 | nan |
| f1 | 0.960352 | 0.482133 |
| accuracy | 0.960699 | 0.482133 |
| precision | 0.968889 | 0.482133 |
| recall | 0.951965 | 0.482133 |
| mcc | 0.921538 | 0.482133 |
Confusion matrix (at threshold=0.482133)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 222 | 7 |
| Labeled as 1 | 11 | 218 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 17_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.1
- depth: 7
- rsm: 1.0
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
28.4 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.2624 | nan |
| auc | 0.98321 | nan |
| f1 | 0.964758 | 0.476858 |
| accuracy | 0.965066 | 0.476858 |
| precision | 1 | 0.853906 |
| recall | 1 | 0.0101198 |
| mcc | 0.930699 | 0.50387 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.2624 | nan |
| auc | 0.98321 | nan |
| f1 | 0.964758 | 0.476858 |
| accuracy | 0.965066 | 0.476858 |
| precision | 0.973333 | 0.476858 |
| recall | 0.956332 | 0.476858 |
| mcc | 0.930273 | 0.476858 |
Confusion matrix (at threshold=0.476858)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 223 | 6 |
| Labeled as 1 | 10 | 219 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 17_CatBoost_GoldenFeatures
CatBoost
- n_jobs: -1
- learning_rate: 0.1
- depth: 7
- rsm: 1.0
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.3 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.212563 | nan |
| auc | 0.982847 | nan |
| f1 | 0.956332 | 0.480383 |
| accuracy | 0.956332 | 0.480383 |
| precision | 1 | 0.842453 |
| recall | 1 | 0.0150311 |
| mcc | 0.912803 | 0.523701 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.212563 | nan |
| auc | 0.982847 | nan |
| f1 | 0.956332 | 0.480383 |
| accuracy | 0.956332 | 0.480383 |
| precision | 0.956332 | 0.480383 |
| recall | 0.956332 | 0.480383 |
| mcc | 0.912664 | 0.480383 |
Confusion matrix (at threshold=0.480383)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 219 | 10 |
| Labeled as 1 | 10 | 219 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 18_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.2
- depth: 8
- rsm: 1.0
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.2 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.200394 | nan |
| auc | 0.985107 | nan |
| f1 | 0.964758 | 0.502573 |
| accuracy | 0.965066 | 0.502573 |
| precision | 1 | 0.854009 |
| recall | 1 | 0.00343254 |
| mcc | 0.930273 | 0.502573 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.200394 | nan |
| auc | 0.985107 | nan |
| f1 | 0.964758 | 0.502573 |
| accuracy | 0.965066 | 0.502573 |
| precision | 0.973333 | 0.502573 |
| recall | 0.956332 | 0.502573 |
| mcc | 0.930273 | 0.502573 |
Confusion matrix (at threshold=0.502573)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 223 | 6 |
| Labeled as 1 | 10 | 219 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 19_RandomForest
Random Forest
- n_jobs: -1
- criterion: gini
- max_features: 0.5
- min_samples_split: 20
- max_depth: 4
- eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.4 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.209104 | nan |
| auc | 0.980826 | nan |
| f1 | 0.955947 | 0.484429 |
| accuracy | 0.956332 | 0.484429 |
| precision | 1 | 0.945715 |
| recall | 1 | 0.0191015 |
| mcc | 0.913221 | 0.510067 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.209104 | nan |
| auc | 0.980826 | nan |
| f1 | 0.955947 | 0.484429 |
| accuracy | 0.956332 | 0.484429 |
| precision | 0.964444 | 0.484429 |
| recall | 0.947598 | 0.484429 |
| mcc | 0.912803 | 0.484429 |
Confusion matrix (at threshold=0.484429)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 221 | 8 |
| Labeled as 1 | 12 | 217 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 1_Linear
Logistic Regression (Linear)
- n_jobs: -1
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.5 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.192042 | nan |
| auc | 0.974028 | nan |
| f1 | 0.941176 | 0.538017 |
| accuracy | 0.943231 | 0.538017 |
| precision | 1 | 0.88594 |
| recall | 1 | 0.00697453 |
| mcc | 0.888635 | 0.538017 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.192042 | nan |
| auc | 0.974028 | nan |
| f1 | 0.941176 | 0.538017 |
| accuracy | 0.943231 | 0.538017 |
| precision | 0.976526 | 0.538017 |
| recall | 0.908297 | 0.538017 |
| mcc | 0.888635 | 0.538017 |
Confusion matrix (at threshold=0.538017)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 224 | 5 |
| Labeled as 1 | 21 | 208 |
Learning curves
Coefficients
| feature | Learner_1 | Learner_2 | Learner_3 | Learner_4 | Learner_5 |
|---|---|---|---|---|---|
| PEER_PRESSURE | 2.09855 | 1.97751 | 2.10596 | 2.24214 | 2.11845 |
| ALLERGY | 1.71851 | 1.8115 | 1.80494 | 1.6434 | 1.99987 |
| CHRONIC DISEASE | 1.87657 | 1.63242 | 1.72145 | 1.71654 | 1.80335 |
| YELLOW_FINGERS | 1.60769 | 1.59401 | 1.57961 | 1.70178 | 1.69653 |
| SWALLOWING DIFFICULTY | 1.54115 | 1.39416 | 1.52935 | 1.57504 | 1.50732 |
| COUGHING | 1.6269 | 1.44695 | 1.61924 | 1.30983 | 1.45889 |
| WHEEZING | 1.39098 | 1.33944 | 1.35351 | 1.6218 | 1.23305 |
| ALCOHOL CONSUMING | 1.2172 | 1.51097 | 1.01974 | 1.43756 | 1.6208 |
| FATIGUE | 1.22949 | 1.23666 | 1.24562 | 1.05542 | 1.27239 |
| ANXIETY | 0.676465 | 1.11655 | 0.820949 | 1.03973 | 0.927658 |
| CHEST PAIN | 0.451082 | 0.609851 | 0.301484 | 0.389733 | 0.68452 |
| SMOKING | 0.404751 | 0.352542 | 0.398707 | 0.489795 | 0.317827 |
| SHORTNESS OF BREATH | 0.222579 | 0.360146 | 0.254631 | 0.399099 | 0.113896 |
| AGE | -0.0683001 | 0.0651653 | 0.0293103 | -0.0196802 | 0.14053 |
| GENDER | -0.0705105 | -0.180772 | -0.240995 | -0.128562 | 0.147742 |
| intercept | -4.88958 | -5.06286 | -4.78165 | -5.16035 | -5.28215 |
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 20_RandomForest
Random Forest
- n_jobs: -1
- criterion: gini
- max_features: 0.7
- min_samples_split: 30
- max_depth: 7
- eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.4 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.241247 | nan |
| auc | 0.976402 | nan |
| f1 | 0.922747 | 0.48316 |
| accuracy | 0.921397 | 0.48316 |
| precision | 1 | 0.70647 |
| recall | 1 | 0 |
| mcc | 0.84331 | 0.48316 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.241247 | nan |
| auc | 0.976402 | nan |
| f1 | 0.922747 | 0.48316 |
| accuracy | 0.921397 | 0.48316 |
| precision | 0.907173 | 0.48316 |
| recall | 0.938865 | 0.48316 |
| mcc | 0.84331 | 0.48316 |
Confusion matrix (at threshold=0.48316)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 207 | 22 |
| Labeled as 1 | 14 | 215 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 21_RandomForest
Random Forest
- n_jobs: -1
- criterion: gini
- max_features: 0.8
- min_samples_split: 40
- max_depth: 3
- eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.1 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.299401 | nan |
| auc | 0.945415 | nan |
| f1 | 0.904564 | 0.497505 |
| accuracy | 0.899563 | 0.497505 |
| precision | 1 | 0.796151 |
| recall | 1 | 0.0269788 |
| mcc | 0.803552 | 0.497505 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.299401 | nan |
| auc | 0.945415 | nan |
| f1 | 0.904564 | 0.497505 |
| accuracy | 0.899563 | 0.497505 |
| precision | 0.86166 | 0.497505 |
| recall | 0.951965 | 0.497505 |
| mcc | 0.803552 | 0.497505 |
Confusion matrix (at threshold=0.497505)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 194 | 35 |
| Labeled as 1 | 11 | 218 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 22_RandomForest
Random Forest
- n_jobs: -1
- criterion: entropy
- max_features: 0.6
- min_samples_split: 50
- max_depth: 6
- eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.8 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.325746 | nan |
| auc | 0.955426 | nan |
| f1 | 0.918919 | 0.498894 |
| accuracy | 0.914847 | 0.498894 |
| precision | 1 | 0.8082 |
| recall | 0.995633 | 0 |
| mcc | 0.833911 | 0.498894 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.325746 | nan |
| auc | 0.955426 | nan |
| f1 | 0.918919 | 0.498894 |
| accuracy | 0.914847 | 0.498894 |
| precision | 0.876984 | 0.498894 |
| recall | 0.965066 | 0.498894 |
| mcc | 0.833911 | 0.498894 |
Confusion matrix (at threshold=0.498894)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 198 | 31 |
| Labeled as 1 | 8 | 221 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 23_NeuralNetwork
Neural Network
- n_jobs: -1
- dense_1_size: 32
- dense_2_size: 4
- learning_rate: 0.05
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
27.0 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.232964 | nan |
| auc | 0.976917 | nan |
| f1 | 0.944321 | 0.492837 |
| accuracy | 0.945415 | 0.492837 |
| precision | 1 | 0.99338 |
| recall | 1 | 6.41845e-12 |
| mcc | 0.891518 | 0.492837 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.232964 | nan |
| auc | 0.976917 | nan |
| f1 | 0.944321 | 0.492837 |
| accuracy | 0.945415 | 0.492837 |
| precision | 0.963636 | 0.492837 |
| recall | 0.925764 | 0.492837 |
| mcc | 0.891518 | 0.492837 |
Confusion matrix (at threshold=0.492837)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 221 | 8 |
| Labeled as 1 | 17 | 212 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 24_NeuralNetwork
Neural Network
- n_jobs: -1
- dense_1_size: 64
- dense_2_size: 16
- learning_rate: 0.01
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
27.6 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.192402 | nan |
| auc | 0.974066 | nan |
| f1 | 0.944321 | 0.509772 |
| accuracy | 0.945415 | 0.509772 |
| precision | 1 | 0.999026 |
| recall | 1 | 0.000268946 |
| mcc | 0.891518 | 0.509772 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.192402 | nan |
| auc | 0.974066 | nan |
| f1 | 0.944321 | 0.509772 |
| accuracy | 0.945415 | 0.509772 |
| precision | 0.963636 | 0.509772 |
| recall | 0.925764 | 0.509772 |
| mcc | 0.891518 | 0.509772 |
Confusion matrix (at threshold=0.509772)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 221 | 8 |
| Labeled as 1 | 17 | 212 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 25_NeuralNetwork
Neural Network
- n_jobs: -1
- dense_1_size: 64
- dense_2_size: 32
- learning_rate: 0.08
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
28.3 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.244839 | nan |
| auc | 0.966877 | nan |
| f1 | 0.938865 | 0.505872 |
| accuracy | 0.938865 | 0.505872 |
| precision | 1 | 0.991168 |
| recall | 1 | 4.76261e-06 |
| mcc | 0.877863 | 0.551479 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.244839 | nan |
| auc | 0.966877 | nan |
| f1 | 0.938865 | 0.505872 |
| accuracy | 0.938865 | 0.505872 |
| precision | 0.938865 | 0.505872 |
| recall | 0.938865 | 0.505872 |
| mcc | 0.877729 | 0.505872 |
Confusion matrix (at threshold=0.505872)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 215 | 14 |
| Labeled as 1 | 14 | 215 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 26_NeuralNetwork
Neural Network
- n_jobs: -1
- dense_1_size: 32
- dense_2_size: 32
- learning_rate: 0.05
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
28.1 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.190663 | nan |
| auc | 0.979119 | nan |
| f1 | 0.947598 | 0.46477 |
| accuracy | 0.947598 | 0.46477 |
| precision | 1 | 0.999892 |
| recall | 1 | 4.76882e-06 |
| mcc | 0.895197 | 0.46477 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.190663 | nan |
| auc | 0.979119 | nan |
| f1 | 0.947598 | 0.46477 |
| accuracy | 0.947598 | 0.46477 |
| precision | 0.947598 | 0.46477 |
| recall | 0.947598 | 0.46477 |
| mcc | 0.895197 | 0.46477 |
Confusion matrix (at threshold=0.46477)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 217 | 12 |
| Labeled as 1 | 12 | 217 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 27_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 95
- learning_rate: 0.05
- feature_fraction: 0.9
- bagging_fraction: 1.0
- min_data_in_leaf: 10
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.6 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.185771 | nan |
| auc | 0.977575 | nan |
| f1 | 0.960352 | 0.468831 |
| accuracy | 0.960699 | 0.468831 |
| precision | 1 | 0.95207 |
| recall | 1 | 0.0219251 |
| mcc | 0.92196 | 0.508432 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.185771 | nan |
| auc | 0.977575 | nan |
| f1 | 0.960352 | 0.468831 |
| accuracy | 0.960699 | 0.468831 |
| precision | 0.968889 | 0.468831 |
| recall | 0.951965 | 0.468831 |
| mcc | 0.921538 | 0.468831 |
Confusion matrix (at threshold=0.468831)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 222 | 7 |
| Labeled as 1 | 11 | 218 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 28_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.1
- depth: 7
- rsm: 0.9
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.8 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.274684 | nan |
| auc | 0.982857 | nan |
| f1 | 0.964444 | 0.50655 |
| accuracy | 0.965066 | 0.50655 |
| precision | 0.990826 | 0.774811 |
| recall | 1 | 0.00625214 |
| mcc | 0.930699 | 0.50655 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.274684 | nan |
| auc | 0.982857 | nan |
| f1 | 0.964444 | 0.50655 |
| accuracy | 0.965066 | 0.50655 |
| precision | 0.9819 | 0.50655 |
| recall | 0.947598 | 0.50655 |
| mcc | 0.930699 | 0.50655 |
Confusion matrix (at threshold=0.50655)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 225 | 4 |
| Labeled as 1 | 12 | 217 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 29_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 63
- learning_rate: 0.05
- feature_fraction: 0.8
- bagging_fraction: 1.0
- min_data_in_leaf: 20
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.3 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.2393 | nan |
| auc | 0.980921 | nan |
| f1 | 0.964758 | 0.49049 |
| accuracy | 0.965066 | 0.49049 |
| precision | 1 | 0.887579 |
| recall | 1 | 0.00673189 |
| mcc | 0.930273 | 0.49049 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.2393 | nan |
| auc | 0.980921 | nan |
| f1 | 0.964758 | 0.49049 |
| accuracy | 0.965066 | 0.49049 |
| precision | 0.973333 | 0.49049 |
| recall | 0.956332 | 0.49049 |
| mcc | 0.930273 | 0.49049 |
Confusion matrix (at threshold=0.49049)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 223 | 6 |
| Labeled as 1 | 10 | 219 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 2_Default_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 63
- learning_rate: 0.05
- feature_fraction: 0.9
- bagging_fraction: 0.9
- min_data_in_leaf: 10
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
26.7 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.224347 | nan |
| auc | 0.975811 | nan |
| f1 | 0.962138 | 0.496641 |
| accuracy | 0.962882 | 0.496641 |
| precision | 0.993939 | 0.751429 |
| recall | 1 | 0.016086 |
| mcc | 0.92648 | 0.496641 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.224347 | nan |
| auc | 0.975811 | nan |
| f1 | 0.962138 | 0.496641 |
| accuracy | 0.962882 | 0.496641 |
| precision | 0.981818 | 0.496641 |
| recall | 0.943231 | 0.496641 |
| mcc | 0.92648 | 0.496641 |
Confusion matrix (at threshold=0.496641)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 225 | 4 |
| Labeled as 1 | 13 | 216 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 30_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 63
- learning_rate: 0.05
- feature_fraction: 1.0
- bagging_fraction: 1.0
- min_data_in_leaf: 20
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.6 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.230054 | nan |
| auc | 0.979958 | nan |
| f1 | 0.969163 | 0.499229 |
| accuracy | 0.969432 | 0.499229 |
| precision | 0.99422 | 0.771052 |
| recall | 1 | 0.00285851 |
| mcc | 0.939008 | 0.499229 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.230054 | nan |
| auc | 0.979958 | nan |
| f1 | 0.969163 | 0.499229 |
| accuracy | 0.969432 | 0.499229 |
| precision | 0.977778 | 0.499229 |
| recall | 0.960699 | 0.499229 |
| mcc | 0.939008 | 0.499229 |
Confusion matrix (at threshold=0.499229)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 224 | 5 |
| Labeled as 1 | 9 | 220 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 31_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.1
- depth: 6
- rsm: 0.9
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.3 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.270426 | nan |
| auc | 0.978624 | nan |
| f1 | 0.967177 | 0.486433 |
| accuracy | 0.967249 | 0.486433 |
| precision | 1 | 0.895824 |
| recall | 1 | 0.0031013 |
| mcc | 0.934507 | 0.486433 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.270426 | nan |
| auc | 0.978624 | nan |
| f1 | 0.967177 | 0.486433 |
| accuracy | 0.967249 | 0.486433 |
| precision | 0.969298 | 0.486433 |
| recall | 0.965066 | 0.486433 |
| mcc | 0.934507 | 0.486433 |
Confusion matrix (at threshold=0.486433)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 222 | 7 |
| Labeled as 1 | 8 | 221 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 32_Xgboost
Extreme Gradient Boosting (Xgboost)
- n_jobs: -1
- objective: binary:logistic
- eta: 0.075
- max_depth: 6
- min_child_weight: 5
- subsample: 1.0
- colsample_bytree: 1.0
- eval_metric: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.2 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.2715 | nan |
| auc | 0.972054 | nan |
| f1 | 0.947137 | 0.494361 |
| accuracy | 0.947598 | 0.494361 |
| precision | 1 | 0.928547 |
| recall | 1 | 0.0124766 |
| mcc | 0.895333 | 0.494361 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.2715 | nan |
| auc | 0.972054 | nan |
| f1 | 0.947137 | 0.494361 |
| accuracy | 0.947598 | 0.494361 |
| precision | 0.955556 | 0.494361 |
| recall | 0.938865 | 0.494361 |
| mcc | 0.895333 | 0.494361 |
Confusion matrix (at threshold=0.494361)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 219 | 10 |
| Labeled as 1 | 14 | 215 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 33_RandomForest
Random Forest
- n_jobs: -1
- criterion: gini
- max_features: 0.5
- min_samples_split: 10
- max_depth: 4
- eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
31.6 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.200104 | nan |
| auc | 0.978833 | nan |
| f1 | 0.955947 | 0.470352 |
| accuracy | 0.956332 | 0.470352 |
| precision | 1 | 0.902146 |
| recall | 1 | 0.0139463 |
| mcc | 0.913221 | 0.510375 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.200104 | nan |
| auc | 0.978833 | nan |
| f1 | 0.955947 | 0.470352 |
| accuracy | 0.956332 | 0.470352 |
| precision | 0.964444 | 0.470352 |
| recall | 0.947598 | 0.470352 |
| mcc | 0.912803 | 0.470352 |
Confusion matrix (at threshold=0.470352)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 221 | 8 |
| Labeled as 1 | 12 | 217 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 34_RandomForest
Random Forest
- n_jobs: -1
- criterion: gini
- max_features: 0.5
- min_samples_split: 30
- max_depth: 4
- eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
32.0 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.27775 | nan |
| auc | 0.972331 | nan |
| f1 | 0.939394 | 0.502531 |
| accuracy | 0.938865 | 0.502531 |
| precision | 0.988166 | 0.749075 |
| recall | 1 | 0.015909 |
| mcc | 0.877863 | 0.502531 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.27775 | nan |
| auc | 0.972331 | nan |
| f1 | 0.939394 | 0.502531 |
| accuracy | 0.938865 | 0.502531 |
| precision | 0.93133 | 0.502531 |
| recall | 0.947598 | 0.502531 |
| mcc | 0.877863 | 0.502531 |
Confusion matrix (at threshold=0.502531)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 213 | 16 |
| Labeled as 1 | 12 | 217 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 35_Xgboost
Extreme Gradient Boosting (Xgboost)
- n_jobs: -1
- objective: binary:logistic
- eta: 0.1
- max_depth: 8
- min_child_weight: 5
- subsample: 1.0
- colsample_bytree: 1.0
- eval_metric: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
31.0 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.264819 | nan |
| auc | 0.975763 | nan |
| f1 | 0.951542 | 0.498066 |
| accuracy | 0.951965 | 0.498066 |
| precision | 1 | 0.927489 |
| recall | 1 | 0.00842946 |
| mcc | 0.904068 | 0.498066 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.264819 | nan |
| auc | 0.975763 | nan |
| f1 | 0.951542 | 0.498066 |
| accuracy | 0.951965 | 0.498066 |
| precision | 0.96 | 0.498066 |
| recall | 0.943231 | 0.498066 |
| mcc | 0.904068 | 0.498066 |
Confusion matrix (at threshold=0.498066)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 220 | 9 |
| Labeled as 1 | 13 | 216 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 36_NeuralNetwork
Neural Network
- n_jobs: -1
- dense_1_size: 32
- dense_2_size: 32
- learning_rate: 0.01
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.8 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.24037 | nan |
| auc | 0.973742 | nan |
| f1 | 0.926407 | 0.540293 |
| accuracy | 0.925764 | 0.540293 |
| precision | 1 | 0.983783 |
| recall | 1 | 1.60164e-05 |
| mcc | 0.851658 | 0.540293 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.24037 | nan |
| auc | 0.973742 | nan |
| f1 | 0.926407 | 0.540293 |
| accuracy | 0.925764 | 0.540293 |
| precision | 0.918455 | 0.540293 |
| recall | 0.934498 | 0.540293 |
| mcc | 0.851658 | 0.540293 |
Confusion matrix (at threshold=0.540293)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 210 | 19 |
| Labeled as 1 | 15 | 214 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 37_NeuralNetwork
Neural Network
- n_jobs: -1
- dense_1_size: 32
- dense_2_size: 32
- learning_rate: 0.08
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.0 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.231553 | nan |
| auc | 0.974829 | nan |
| f1 | 0.943231 | 0.629602 |
| accuracy | 0.943231 | 0.629602 |
| precision | 1 | 0.999927 |
| recall | 1 | 1.97928e-06 |
| mcc | 0.886463 | 0.629602 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.231553 | nan |
| auc | 0.974829 | nan |
| f1 | 0.943231 | 0.629602 |
| accuracy | 0.943231 | 0.629602 |
| precision | 0.943231 | 0.629602 |
| recall | 0.943231 | 0.629602 |
| mcc | 0.886463 | 0.629602 |
Confusion matrix (at threshold=0.629602)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 216 | 13 |
| Labeled as 1 | 13 | 216 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 38_NeuralNetwork
Neural Network
- n_jobs: -1
- dense_1_size: 32
- dense_2_size: 4
- learning_rate: 0.01
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.4 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.170013 | nan |
| auc | 0.982418 | nan |
| f1 | 0.941704 | 0.624584 |
| accuracy | 0.943231 | 0.624584 |
| precision | 1 | 0.99965 |
| recall | 1 | 4.58067e-09 |
| mcc | 0.887682 | 0.624584 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.170013 | nan |
| auc | 0.982418 | nan |
| f1 | 0.941704 | 0.624584 |
| accuracy | 0.943231 | 0.624584 |
| precision | 0.967742 | 0.624584 |
| recall | 0.917031 | 0.624584 |
| mcc | 0.887682 | 0.624584 |
Confusion matrix (at threshold=0.624584)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 222 | 7 |
| Labeled as 1 | 19 | 210 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 39_NeuralNetwork
Neural Network
- n_jobs: -1
- dense_1_size: 32
- dense_2_size: 4
- learning_rate: 0.08
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.9 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.272124 | nan |
| auc | 0.970729 | nan |
| f1 | 0.9375 | 0.494036 |
| accuracy | 0.938865 | 0.494036 |
| precision | 0.99115 | 0.992942 |
| recall | 1 | 1.50593e-14 |
| mcc | 0.878567 | 0.494036 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.272124 | nan |
| auc | 0.970729 | nan |
| f1 | 0.9375 | 0.494036 |
| accuracy | 0.938865 | 0.494036 |
| precision | 0.958904 | 0.494036 |
| recall | 0.917031 | 0.494036 |
| mcc | 0.878567 | 0.494036 |
Confusion matrix (at threshold=0.494036)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 220 | 9 |
| Labeled as 1 | 19 | 210 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 3_Default_Xgboost
Extreme Gradient Boosting (Xgboost)
- n_jobs: -1
- objective: binary:logistic
- eta: 0.075
- max_depth: 6
- min_child_weight: 1
- subsample: 1.0
- colsample_bytree: 1.0
- eval_metric: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
26.9 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.221408 | nan |
| auc | 0.980082 | nan |
| f1 | 0.96 | 0.499264 |
| accuracy | 0.960699 | 0.499264 |
| precision | 1 | 0.896944 |
| recall | 1 | 0.0151183 |
| mcc | 0.92196 | 0.499264 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.221408 | nan |
| auc | 0.980082 | nan |
| f1 | 0.96 | 0.499264 |
| accuracy | 0.960699 | 0.499264 |
| precision | 0.977376 | 0.499264 |
| recall | 0.943231 | 0.499264 |
| mcc | 0.92196 | 0.499264 |
Confusion matrix (at threshold=0.499264)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 224 | 5 |
| Labeled as 1 | 13 | 216 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 40_RandomForest
Random Forest
- n_jobs: -1
- criterion: gini
- max_features: 0.7
- min_samples_split: 20
- max_depth: 7
- eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
33.2 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.238848 | nan |
| auc | 0.98177 | nan |
| f1 | 0.96 | 0.515969 |
| accuracy | 0.960699 | 0.515969 |
| precision | 1 | 0.823299 |
| recall | 0.995633 | 0 |
| mcc | 0.92196 | 0.515969 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.238848 | nan |
| auc | 0.98177 | nan |
| f1 | 0.96 | 0.515969 |
| accuracy | 0.960699 | 0.515969 |
| precision | 0.977376 | 0.515969 |
| recall | 0.943231 | 0.515969 |
| mcc | 0.92196 | 0.515969 |
Confusion matrix (at threshold=0.515969)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 224 | 5 |
| Labeled as 1 | 13 | 216 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 41_RandomForest
Random Forest
- n_jobs: -1
- criterion: gini
- max_features: 0.7
- min_samples_split: 40
- max_depth: 7
- eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
32.5 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.302278 | nan |
| auc | 0.969699 | nan |
| f1 | 0.916844 | 0.5789 |
| accuracy | 0.917031 | 0.614978 |
| precision | 1 | 0.700223 |
| recall | 0.995633 | 0 |
| mcc | 0.834093 | 0.614978 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.302278 | nan |
| auc | 0.969699 | nan |
| f1 | 0.916667 | 0.614978 |
| accuracy | 0.917031 | 0.614978 |
| precision | 0.920705 | 0.614978 |
| recall | 0.912664 | 0.614978 |
| mcc | 0.834093 | 0.614978 |
Confusion matrix (at threshold=0.614978)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 211 | 18 |
| Labeled as 1 | 20 | 209 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 42_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.1
- depth: 6
- rsm: 0.8
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
31.9 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.273489 | nan |
| auc | 0.985622 | nan |
| f1 | 0.969163 | 0.486184 |
| accuracy | 0.969432 | 0.486184 |
| precision | 1 | 0.721253 |
| recall | 1 | 0.0074462 |
| mcc | 0.939008 | 0.486184 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.273489 | nan |
| auc | 0.985622 | nan |
| f1 | 0.969163 | 0.486184 |
| accuracy | 0.969432 | 0.486184 |
| precision | 0.977778 | 0.486184 |
| recall | 0.960699 | 0.486184 |
| mcc | 0.939008 | 0.486184 |
Confusion matrix (at threshold=0.486184)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 224 | 5 |
| Labeled as 1 | 9 | 220 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 43_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 63
- learning_rate: 0.1
- feature_fraction: 1.0
- bagging_fraction: 1.0
- min_data_in_leaf: 20
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
31.1 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.240681 | nan |
| auc | 0.978395 | nan |
| f1 | 0.966887 | 0.498922 |
| accuracy | 0.967249 | 0.498922 |
| precision | 0.992248 | 0.85133 |
| recall | 1 | 0.00219441 |
| mcc | 0.934721 | 0.498922 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.240681 | nan |
| auc | 0.978395 | nan |
| f1 | 0.966887 | 0.498922 |
| accuracy | 0.967249 | 0.498922 |
| precision | 0.977679 | 0.498922 |
| recall | 0.956332 | 0.498922 |
| mcc | 0.934721 | 0.498922 |
Confusion matrix (at threshold=0.498922)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 224 | 5 |
| Labeled as 1 | 10 | 219 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 44_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.1
- depth: 7
- rsm: 0.8
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.8 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.30591 | nan |
| auc | 0.984249 | nan |
| f1 | 0.969432 | 0.44434 |
| accuracy | 0.969432 | 0.44434 |
| precision | 1 | 0.674742 |
| recall | 1 | 0.0236459 |
| mcc | 0.938865 | 0.44434 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.30591 | nan |
| auc | 0.984249 | nan |
| f1 | 0.969432 | 0.44434 |
| accuracy | 0.969432 | 0.44434 |
| precision | 0.969432 | 0.44434 |
| recall | 0.969432 | 0.44434 |
| mcc | 0.938865 | 0.44434 |
Confusion matrix (at threshold=0.44434)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 222 | 7 |
| Labeled as 1 | 7 | 222 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 45_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.1
- depth: 7
- rsm: 1
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.8 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.2624 | nan |
| auc | 0.98321 | nan |
| f1 | 0.964758 | 0.476858 |
| accuracy | 0.965066 | 0.476858 |
| precision | 1 | 0.853906 |
| recall | 1 | 0.0101198 |
| mcc | 0.930699 | 0.50387 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.2624 | nan |
| auc | 0.98321 | nan |
| f1 | 0.964758 | 0.476858 |
| accuracy | 0.965066 | 0.476858 |
| precision | 0.973333 | 0.476858 |
| recall | 0.956332 | 0.476858 |
| mcc | 0.930273 | 0.476858 |
Confusion matrix (at threshold=0.476858)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 223 | 6 |
| Labeled as 1 | 10 | 219 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 46_LightGBM
LightGBM
- n_jobs: -1
- objective: binary
- num_leaves: 63
- learning_rate: 0.1
- feature_fraction: 0.5
- bagging_fraction: 1.0
- min_data_in_leaf: 10
- metric: custom
- custom_eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
30.9 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.382829 | nan |
| auc | 0.977146 | nan |
| f1 | 0.964758 | 0.502209 |
| accuracy | 0.965066 | 0.502209 |
| precision | 1 | 0.713541 |
| recall | 1 | 0.145873 |
| mcc | 0.930273 | 0.502209 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.382829 | nan |
| auc | 0.977146 | nan |
| f1 | 0.964758 | 0.502209 |
| accuracy | 0.965066 | 0.502209 |
| precision | 0.973333 | 0.502209 |
| recall | 0.956332 | 0.502209 |
| mcc | 0.930273 | 0.502209 |
Confusion matrix (at threshold=0.502209)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 223 | 6 |
| Labeled as 1 | 10 | 219 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 47_Xgboost
Extreme Gradient Boosting (Xgboost)
- n_jobs: -1
- objective: binary:logistic
- eta: 0.075
- max_depth: 5
- min_child_weight: 1
- subsample: 1.0
- colsample_bytree: 1.0
- eval_metric: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
31.8 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.167821 | nan |
| auc | 0.981684 | nan |
| f1 | 0.96 | 0.506348 |
| accuracy | 0.960699 | 0.506348 |
| precision | 1 | 0.962576 |
| recall | 1 | 0.0105037 |
| mcc | 0.92196 | 0.506348 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.167821 | nan |
| auc | 0.981684 | nan |
| f1 | 0.96 | 0.506348 |
| accuracy | 0.960699 | 0.506348 |
| precision | 0.977376 | 0.506348 |
| recall | 0.943231 | 0.506348 |
| mcc | 0.92196 | 0.506348 |
Confusion matrix (at threshold=0.506348)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 224 | 5 |
| Labeled as 1 | 13 | 216 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 4_Default_CatBoost
CatBoost
- n_jobs: -1
- learning_rate: 0.1
- depth: 6
- rsm: 1
- loss_function: Logloss
- eval_metric: Accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
26.6 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.193208 | nan |
| auc | 0.984421 | nan |
| f1 | 0.960352 | 0.485147 |
| accuracy | 0.960699 | 0.485147 |
| precision | 1 | 0.955622 |
| recall | 1 | 0.00483837 |
| mcc | 0.92196 | 0.51319 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.193208 | nan |
| auc | 0.984421 | nan |
| f1 | 0.960352 | 0.485147 |
| accuracy | 0.960699 | 0.485147 |
| precision | 0.968889 | 0.485147 |
| recall | 0.951965 | 0.485147 |
| mcc | 0.921538 | 0.485147 |
Confusion matrix (at threshold=0.485147)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 222 | 7 |
| Labeled as 1 | 11 | 218 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 5_Default_NeuralNetwork
Neural Network
- n_jobs: -1
- dense_1_size: 32
- dense_2_size: 16
- learning_rate: 0.05
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
25.2 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.261498 | nan |
| auc | 0.969814 | nan |
| f1 | 0.942731 | 0.422963 |
| accuracy | 0.943231 | 0.422963 |
| precision | 1 | 0.999584 |
| recall | 1 | 2.82294e-13 |
| mcc | 0.886598 | 0.422963 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.261498 | nan |
| auc | 0.969814 | nan |
| f1 | 0.942731 | 0.422963 |
| accuracy | 0.943231 | 0.422963 |
| precision | 0.951111 | 0.422963 |
| recall | 0.934498 | 0.422963 |
| mcc | 0.886598 | 0.422963 |
Confusion matrix (at threshold=0.422963)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 218 | 11 |
| Labeled as 1 | 15 | 214 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 6_Default_RandomForest
Random Forest
- n_jobs: -1
- criterion: gini
- max_features: 0.9
- min_samples_split: 30
- max_depth: 4
- eval_metric_name: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
29.3 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.250484 | nan |
| auc | 0.967792 | nan |
| f1 | 0.921397 | 0.50911 |
| accuracy | 0.921397 | 0.50911 |
| precision | 1 | 0.876451 |
| recall | 1 | 0.0074135 |
| mcc | 0.842795 | 0.50911 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.250484 | nan |
| auc | 0.967792 | nan |
| f1 | 0.921397 | 0.50911 |
| accuracy | 0.921397 | 0.50911 |
| precision | 0.921397 | 0.50911 |
| recall | 0.921397 | 0.50911 |
| mcc | 0.842795 | 0.50911 |
Confusion matrix (at threshold=0.50911)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 211 | 18 |
| Labeled as 1 | 18 | 211 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 7_Xgboost
Extreme Gradient Boosting (Xgboost)
- n_jobs: -1
- objective: binary:logistic
- eta: 0.075
- max_depth: 8
- min_child_weight: 5
- subsample: 1.0
- colsample_bytree: 1.0
- eval_metric: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
27.5 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.2715 | nan |
| auc | 0.972054 | nan |
| f1 | 0.947137 | 0.494361 |
| accuracy | 0.947598 | 0.494361 |
| precision | 1 | 0.928547 |
| recall | 1 | 0.0124766 |
| mcc | 0.895333 | 0.494361 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.2715 | nan |
| auc | 0.972054 | nan |
| f1 | 0.947137 | 0.494361 |
| accuracy | 0.947598 | 0.494361 |
| precision | 0.955556 | 0.494361 |
| recall | 0.938865 | 0.494361 |
| mcc | 0.895333 | 0.494361 |
Confusion matrix (at threshold=0.494361)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 219 | 10 |
| Labeled as 1 | 14 | 215 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 8_Xgboost
Extreme Gradient Boosting (Xgboost)
- n_jobs: -1
- objective: binary:logistic
- eta: 0.1
- max_depth: 8
- min_child_weight: 1
- subsample: 1.0
- colsample_bytree: 1.0
- eval_metric: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
27.9 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.167899 | nan |
| auc | 0.982256 | nan |
| f1 | 0.955947 | 0.430063 |
| accuracy | 0.956332 | 0.430063 |
| precision | 1 | 0.919392 |
| recall | 1 | 0.00517778 |
| mcc | 0.913221 | 0.505432 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.167899 | nan |
| auc | 0.982256 | nan |
| f1 | 0.955947 | 0.430063 |
| accuracy | 0.956332 | 0.430063 |
| precision | 0.964444 | 0.430063 |
| recall | 0.947598 | 0.430063 |
| mcc | 0.912803 | 0.430063 |
Confusion matrix (at threshold=0.430063)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 221 | 8 |
| Labeled as 1 | 12 | 217 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of 9_Xgboost
Extreme Gradient Boosting (Xgboost)
- n_jobs: -1
- objective: binary:logistic
- eta: 0.1
- max_depth: 7
- min_child_weight: 25
- subsample: 0.9
- colsample_bytree: 0.6
- eval_metric: accuracy
- explain_level: 1
Validation
- validation_type: kfold
- k_folds: 5
- shuffle: True
- stratify: True
Optimized metric
accuracy
Training time
27.5 seconds
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.474809 | nan |
| auc | 0.91213 | nan |
| f1 | 0.895323 | 0.491758 |
| accuracy | 0.89738 | 0.491758 |
| precision | 0.927632 | 0.562244 |
| recall | 1 | 0.180071 |
| mcc | 0.795374 | 0.491758 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.474809 | nan |
| auc | 0.91213 | nan |
| f1 | 0.895323 | 0.491758 |
| accuracy | 0.89738 | 0.491758 |
| precision | 0.913636 | 0.491758 |
| recall | 0.877729 | 0.491758 |
| mcc | 0.795374 | 0.491758 |
Confusion matrix (at threshold=0.491758)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 210 | 19 |
| Labeled as 1 | 28 | 201 |
Learning curves
Permutation-based Importance
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
Summary of Ensemble
Ensemble structure
| Model | Weight |
|---|---|
| 13_LightGBM | 1 |
| 30_LightGBM | 2 |
| 31_CatBoost | 1 |
Metric details
| score | threshold | |
|---|---|---|
| logloss | 0.233043 | nan |
| auc | 0.982552 | nan |
| f1 | 0.973568 | 0.502666 |
| accuracy | 0.973799 | 0.502666 |
| precision | 1 | 0.88689 |
| recall | 1 | 0.00253774 |
| mcc | 0.947743 | 0.502666 |
Metric details with threshold from accuracy metric
| score | threshold | |
|---|---|---|
| logloss | 0.233043 | nan |
| auc | 0.982552 | nan |
| f1 | 0.973568 | 0.502666 |
| accuracy | 0.973799 | 0.502666 |
| precision | 0.982222 | 0.502666 |
| recall | 0.965066 | 0.502666 |
| mcc | 0.947743 | 0.502666 |
Confusion matrix (at threshold=0.502666)
| Predicted as 0 | Predicted as 1 | |
|---|---|---|
| Labeled as 0 | 225 | 4 |
| Labeled as 1 | 8 | 221 |
Learning curves
Confusion Matrix
Normalized Confusion Matrix
ROC Curve
Kolmogorov-Smirnov Statistic
Precision-Recall Curve
Calibration Curve
Cumulative Gains Curve
Lift Curve
